This ComfyUI node enhances image management by extracting and embedding metadata, ensuring compatibility with various platforms like Civitai. It targets users of ComfyUI, Stable Diffusion, and related tools who need robust prompt and parameter handling for their generated images.

How It Works

The node leverages the core functionality of the standalone SD Prompt Reader, enabling it to parse metadata from images across multiple formats (PNG, JPEG, WEBP, TXT). It integrates with a "Prompt Saver Node" to embed metadata in a widely compatible A1111 format, crucial for website auto-detection and cross-tool compatibility. For SDXL, it merges text_g and text_l prompts into a single string.

Quick Start & Requirements

• Installation: Recommended via ComfyUI Manager. Manual installation requires cloning the repository with submodules (git clone --recursive) and installing dependencies (pip install -r requirements.txt).
• Prerequisites: ComfyUI, Python.
• Links: SD Prompt Reader

Highlighted Details

• Supports metadata extraction from various UIs including A1111, Easy Diffusion, Fooocus-MRE, NovelAI, InvokeAI, and Draw Things.
• The Prompt Saver Node can calculate and embed checkpoint, VAE, and Lora hashes for Civitai auto-detection.
• Includes a Parameter Generator Node to manage parameters for KSampler and a Batch Loader Node for bulk image processing.

Maintenance & Community

• The project is a subproject of the SD Prompt Reader.
• Credits mention inspiration from ComfyUI Load Image With Metadata, Comfy Image Saver, Stable Diffusion Webui, and rgthree's Comfy Nodes.

Licensing & Compatibility

• License details are not explicitly stated in the README.
• Compatibility with commercial use or closed-source linking is not specified.

Limitations & Caveats

Batch reading with the Batch Loader Node does not update preview images and only displays metadata for the last image. PNG is the only format that supports embedding both metadata and workflow simultaneously; other formats only embed metadata.